我从github上下载发布的代码和数据后按照README.md的步骤是可以跑通的,不知道你是遇到了什么问题?

大佬，你发布的TOE项目上的数据和代码是可以跑通的，你提供的数据集是CADEC英文数据集，我现在想适配中文数据集，在数据集格式一样的前提（如下，Resume-zh）下中文数据集跑不通。
[{"sentence": ["高", "勇", "：", "男", "，", "中", "国", "国", "籍", "，", "无", "境", "外", "居", "留", "权", "，"], "ner": [{"index": [0, 1], "type": "NAME"}, {"index": [5, 6, 7, 8], "type": "CONT"}], "word": [[0, 1], [2], [3], [4], [5, 6], [7, 8], [9], [10], [11, 12], [13, 14, 15], [16]]}]

我的config文件（W2NER中提供的resume-zh.json，增加了 "rounds", "alpha", "cr"）如下，bert_config没有改动：

{
  "dataset": "resume-zh",

  "dist_emb_size": 20,
  "type_emb_size": 20,
  "lstm_hid_size": 512,
  "conv_hid_size": 96,
  "bert_hid_size": 768,
  "biaffine_size": 512,
  "ffnn_hid_size": 288,
  "dilation": [1,2,3],

  "emb_dropout": 0.5,
  "conv_dropout": 0.5,
  "out_dropout": 0.33,

  "epochs":10,
  "batch_size": 8,

  "learning_rate": 1e-3,
  "weight_decay": 0,
  "clip_grad_norm": 5.0,

  "bert_name": "bert-base-chinese",
  "bert_learning_rate": 5e-6,
  "warm_factor": 0.1,

  "use_bert_last_4_layers": true,
  "seed": 123,
  
  "rounds": 3,
  "alpha": 0.5,
  "cr": 80
}

报错如下：

如何快速的使用中文数据集？ #2

Description

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions